中文摘要：

现实的群体决策问题往往是复杂的大群体问题,而群体成员之间由于联系程度、性格、心理、价值观等因素的影响可形成不同的社团组织。社团组织的结构,特别是群体中隐性社团组织的划分及其结构无疑对决策结果有重大影响。基于决策成员之间人际关系网络构成的大规模复杂网络,运用节点相似度的凝聚算法思想,研究出节点赋权网络的社团划分新算法。该算法综合考虑节点属性以及节点在网络中的结构特性,分别反映群体决策中决策个体的知识水平及交际网络,用于识别群体中的隐性组织结构,为模拟群体观点演化过程和结果奠定了基础。

英文摘要：

A real group decision problem is often a complex large group problem, and because of the influence of such factors as personality, psychology, values and connection degree, group members can form different community organizations. The structure of community organizations, especially that of the recessive community organizations, has a significant influence on the decision results. Based on the complex network consisting of relationship between members of the group involved in decision making, this paper uses the agglomerative algorithm idea of nodes similarity to design and verify a community partition algorithm for the node empower network. The algorithm considers the properties and structural characteristics of nodes in the network, reflecting both the individual＇s knowledge and communication network in group decision-making. It can be used to identify the structure of the recessive organization involved in group decision-making, thus contributing to the simulation of group evolution process and the decision results.